How expensive is housing in the US?

An examination of the cost of housing in the US over time.

WarningWork in Progress

I am actively updating this post. Content will change as I work my way through this article.

tl;dr

Things are bad, but not as bad as you might think.

At a national scale,

• From January 1953 to January 2026¹, the price-to-income ratio of a home has increased by 2.85 percentage points (pp) per year on average. As of January 2026, the ratio stands at 4.51², placing it in the 73rd percentile of all observations in this period.
• From April 1971 to January 2026, the mortgage debt-to-income ratio has decreased by 0.14 pp per year on average. As of January 2026, the ratio stands at 0.26³, placing it in the 46th percentile of all observations in this period.

¹ Some of the most recent data used to compute these metrics come from forecasts. See Table 2 for details on the range of the underlying data.

² Meaning the typical house costs the typical household 4.51 \(\times\) their gross annual income.

³ Meaning the typical household would spend 26% of their monthly income on the typical mortgage. This calculation assumes only interest and principal.

intro

Much has been written recently on housing affordability, with much of it being negative⁴. Aside from the tendency of news organizations to produce negative content, I think what is lost in these reports is historical context: where were we, and where are we now? Even when the historical context is present, metrics may be used that, if not interpreted carefully, can be misleading. For instance, longtermtrends.com gives a chart that shows the ratio of the S&P/Case-Shiller Home Price Index to the median household income. This index is used a lot since it’s high-quality and has data that dates back to 1890. However, the problem here is that the Case-Shiller index is a weighted average. The distribution of home values tend to be right skewed, so an average will be greater than the median (the typical house). So, dividing the average by the median produces a price-to-income ratio that is a bit more pessimistic. The typical person is buying the typical house, not the average house.

⁴ Here’s a few: CBS, NBC, and FOX.

In order to provide some clarity on the question of housing affordability, I’m going to look at four metrics in particular:

1. The real⁵ median home price over time.
2. The real median household income over time.
3. The ratio of the typical cost of a house to the typical household income. In this article, I’ll refer to it as the price-to-income ratio.
4. The proportion of a person’s monthly income that would be dedicated to their monthly mortgage. For my analysis, this will be limited to only interest and principal⁶, and will assume a 20% downpayment, so a loan on 80% the value of a house. I’ll refer to it as the debt-to-income ratio in the article.

⁵ 2026 dollars.

⁶ I’m interested in having a better picture of this by incorporating property taxes and home insurance costs if someone could get me the data…

The last two metrics are the most important and incorporate the top two. However, the top two can give us better context in and of themselves. EXPLAIN HOW THESE METRICS WILL BE USED IN THE ANALYSIS. EXPLAIN THE FLOW OF THE DOCUMENT.

analysis

I’m planning to examine this at three different geographic scales:

• national (IN PROGRESS)
• state (NOT STARTED)
• county (NOT STARTED)

national scale

data

The data I’m using is mostly from the FRED. I’ve supplemented some of these series with data extending further back to get a more complete picture. Below are the data series and where I got the supplemental data.

Table 1: Data series IDs, descriptions, and supplemental sourcing for national scale analysis

series id	description	sources & notes
CPIAUCSL	Consumer Price Index for All Urban Consumers: All Items in U.S. City Average	CPI for all urban consumers. This index covers roughly 88% of the total population. See series link for details.
MEHOINUSA646N	Median Household Income in the United States	Supplemental data for 1947-1965 and 1967-1983 sourced from US Census Bureau.
MORTGAGE30US	30-Year Fixed Rate Mortgage Average in the U.S.	Restrictive filters on down payments and credit scores help to minimize outliers. See Freddie Mac FAQs for details.
MSPUS	Median Sales Price of Houses Sold for the United States	Years 1953-1962 supplemented via NAR, Shiller, and FHFA indices.

Let’s take a look at the data below:

Figure 1: National housing and economic indicators over time. Dollars are in nominal terms. Hover over the lines to see specific values.

There are a few things that need to be handled before analysis:

1. The data is captured at different frequencies.
2. The data begins⁷ and ends at different dates, with not all the data is recent.

⁷ The 30-year fixed mortgage rate is the most egregious offender. The reason for this, is that Freddie Mac began surveying lenders on mortgage rates in April 1971, making this the earliest date for which consistent, nationwide data is available.

Table 2: Latest data points retrieved for each national economic series.

series	collection frequency	oldest observation	newest observation
30-year fixed mortgage rate	weekly	April 2, 1971	January 15, 2026
consumer price index	monthly	January 1, 1947	December 1, 2025
median home price	quarterly	January 15, 1953	April 1, 2025
median household income	annually	January 1, 1947	January 1, 2024

To account for (1), I put everything in a monthly frequency. Cubic splines were used for quarterly and annual data, while weekly data was averaged. For (2), I first smoothed each series using LOESS then modeled each series using ARIMA to produce estimates for the most recent date available (January 2026). These estimates should be taken with a grain of salt, though I include \(3 \sigma\) prediction intervals to capture the uncertainty. No retrospective prediction is performed⁸, so metrics that depend on more than one indicator begin at the earliest date available in all metrics.

⁸ There’s too large of a discrepancy between the 30-year fixed mortgage rate and other indicators. Any retrospective predictions that tried to cover that discrepancy would be dominated by uncertainty.

Figure 2: Harmonized and smoothed economic indicators over time. Green line shows the smoothed series. Gray line shows the raw series. Dollars are in nominal terms. Hover over the lines to see specific values.

These smoothed indicators are used to calculate the metrics in the following sections.

metrics

real median home price

Let’s put the nominal median home price in 2026 dollars. I’ll include a linear model to help us understand the trend over time.

Figure 3: Real median home prices (2026 dollars) over time. Monthly forecasts with 3\(\sigma\) prediction interval. Linear model with 3\(\sigma\) confidence interval. Hover over the lines to see the percentile of the observation and ranges for forecasts.

From January 1953 to January 2026, the real median home price has increased by $3,462 per year on average. This translates to roughly 1.57 pp per year. As of January 2026, the real median home price stands at $417,986, placing it in the 90th percentile of all observations in this period.

What complicates this analysis is that the typical house from the 1950s looks much different than the typical house from today. Across the decades, houses look pretty different, whether that be square footage or amenities. To get a visual sense of these differences:

The “rambler” (a ranch house) was the most popular style of home for new builds in the 1950s. The home pictured has 3 beds, 1.5 baths, and is 1,234 square feet (excluding the 1-car garage). Images from Town and Country Homes to Fit Your Budget, 1954 p. 60-61.

The modern farmhouse is considered one of the most popular styles of home for new builds in the 2020s. The home pictured has 3 beds, 2.5 baths, and is 1,959 square feet (excluding the 2-car garage). Images from Architectural Designs.

Figure 4: Visual comparison of popular homes from the 1950s and 2020s.

A few key figures on these differences:

Table 3: Comparison of housing features from 1950s to 2020s

feature	1950’s home	2020’s home
median size	under 1,000 sq ft	over 2,000 sq ft
complete plumbing	66%	99%
air conditioning	under 2%	over 90%

Andrew Latham did a great analysis on housing affordability where he found that the price per square foot of a house (ignoring amenities) from 1978-2023 grew by 20% which works out to roughly 0.44 pp per year which is about 28% the rate of house prices in general.

Of course the problem in all this is you can’t buy a 1950’s new-build today; the option for a smaller, less feature rich house just isn’t available today as it was in the past. Those aren’t the houses being built, so buyers are left with nicer, larger⁹, and more expensive houses.

⁹ Ironically, households have been shrinking over the same time that houses have gotten larger. However, the median square footage of new builds has been decreasing over the last decade.

Fortunately, we are coming out of a bubble which peaked in early 2022, so prices are cheaper today than a few years ago. However, unless housing policy/supply changes, we should expect housing to get more expensive.

real household income

Let’s put the nominal median household income in 2026 dollars. I’ll include a linear model to help us understand the trend over time.

Figure 5: real median household income (2026 dollars) over time. Monthly forecasts with 3\(\sigma\) prediction interval. Linear model with 3\(\sigma\) confidence interval. Hover over the lines to see the percentile of the observation and ranges for forecasts.

From January 1947 to January 2026, the real median household income has increased by $398 per year on average. As of January 2026, the real median household income stands at $92,742, placing it in the 100th percentile of all observations in this period.

There are three periods worth focusing on in Figure 5: Mid 1948 - late 1964, late 1964 - early 2012, and early 2012 - today. These periods are marked by significant growth (4.3 pp/year), little to no growth (0.02 pp/year), and substantial growth (1.88 pp/year), respectively. Let’s examine each of these periods in turn.

Mid 1948 - late 1964

This period saw significant growth over a brief period of time. From June 1948 to December 1964, real median household income grew from $42,496 to $72,636 representing an annual growth of 4.3 pp per year. Post WWII, with Europe and Japan having to rebuild their industrial bases and with the establishment of the Bretton Woods System, the US was poised to become an economic powerhouse, facing almost no industrial competition. By the early 1950s, the US produced nearly half (44.8%) of the world’s manufactured goods (Bairoch 1982, 275). Technological advancements were largely responsible for significant increases to productivity (Kendrick 1961) with much of the productivity gains benefiting the typical worker.

Late 1964 - early 2012

This period saw little to no growth. From December 1964 to January 2012, the real median household income only grew to $73,436 representing an annual growth of 0.02 pp per year. Of course there were highs and lows over this period¹⁰, but overall growth was stagnant. Further analyzing this period becomes nebulous quickly. Analysis by Stephen Rose at the Urban Institute, a left-leaning think tank, have found that when you include all payments (e.g. benefits from employer), from 1979 - 2013 real income grew by roughly 1.1 pp per year. Furthermore, when you break up the analysis by different demographics (age, gender, education, etc.) you see some groups had significant growth while others lagged behind (Rose 2015)¹¹.

¹⁰ The peak income over this period was reached on July 1999 at $81,120 representing a growth of 0.34 pp per year. The lowest income over this period was reached on June 1982 at $69,180 representing a decline of -0.14 pp per year.

¹¹ Groups that saw significant growth: Ages 60+, women, and education level of BA or higher. Groups that lagged in growth: Ages 25-34, men, and education levels below BA.

Early 2012 - today

This period has seen substantial growth. From January 2012 to January 2026, the real median household income grew to $92,742 representing an annual growth of 1.88 pp per year. Post the great recession and pre-covid, the unemployment rate declined to its lowest level in 50 years (Gould 2020); a shrinking labor supply relative to demand forced employers to raise wages to attract and retain staff, helping to drive up wages. Post-covid, a similar story occurred, except the lowest-paid workers (10th percentile) saw the most real wage growth¹², driven by young, non-college workers disproportionately moving away from the hospitality sector into higher-paying jobs (Autor, Dube, and McGrew 2023). This compression at the bottom pushed the overall median household income up.

¹² This was measured over January 2020 to June 2023. Over this same period, higher-income workers (90th percentile) saw a decline in real wages

price-to-income ratio

With the context from the previous two metrics, I’ll examine the two most critical metrics: the price-to-income ratio and the debt-to-income ratio. I’ll examine the former first.

The price-to-income is calculated simply as,

\[\frac{\text{median sales price of a house}}{\text{median household income}}\]

This ratio tells us the sales price of a house as a multiple of the typical household income. We can use nominal dollars to calculate the price-to-income ratio. As before, I’ll include a linear model to help us understand the trend over time.

Figure 6: Ratio of median home price to median household income over time. Monthly forecasts with 3\(\sigma\) prediction interval. Linear model with 3\(\sigma\) confidence interval. Hover over the lines to see the percentile of the observation and ranges for forecasts.

From January 1953 to January 2026, the price-to-income ratio of a home has increased by 2.85 pp per year on average. As of January 2026, the ratio stands at 4.51, placing it in the 73rd percentile of all observations in this period.

FOCUS ANALYSIS ON TWO PERIODS: 1953-1963 AND POST 1963. NEED TO INCLUDE RECESSION BARS IN ALL FIGURES.

From 1953 to late 1963, the price-to-income ratio fell precipitously¹³, from a peak of 4.58 on January 1953 to a trough of 2.67 on October 1963 representing an annual decline of 3.87 pp per year. The decline of price-to-income in this period was driven by strong growth in real household income and slight decline in the real median home price.

¹³ This is probably overstated a bit since due to quality of data from pre-1963. See Table 1 for details.

debt-to-income ratio

Figure 7: Ratio of monthly mortgage to monthly income over time. Monthly forecasts with 3\(\sigma\) prediction interval. Linear model with 3\(\sigma\) confidence interval. Hover over the lines to see the percentile of the observation and ranges for forecasts.

From April 1971 to January 2026, the mortgage debt-to-income ratio has decreased by 2.85 pp per year on average. As of January 2026, the ratio stands at 0.26, placing it in the 46th percentile of all observations in this period.

references

Autor, David, Arindrajit Dube, and Annie McGrew. 2023. “The Unexpected Compression: Competition at Work in the Low Wage Labor Market.” National Bureau of Economic Research.

Bairoch, Paul. 1982. “International Industrialization Levels from 1750 to 1980.” Journal of European Economic History.

Gould, Elise. 2020. “State of Working America Wages 2019.” Economic Policy Institute.

Kendrick, John W. 1961. “Productivity Trends in the United States.”

Rose, Stephen. 2015. “Beyond the Wage Stagnation Story: Better Measures Show America’s Workers Doing Better Than Previously Reported.” Urban Institute.